This project, called Diabetes AutoimmunityStudy in the Young (DAISY) began in July 1993 and allowed us to establish two unique cohorts of very young children who are at up to 20-fold increased risk of type 1 diabetes: 1) a cohort of 693 siblings and offspring (current median age 4.5 years) of persons with type 1 diabetes and 1007 of their relatives;and 2) a cohort of 1,069newborns (current median age 3.1 years) with type 1 diabetes-associated HLA-DR,DQ alleles, identified through a cord blood screening of 21,713 general population children without a diabetic relative and 1,491 of their relatives. To date, a short prospective follow-up of these cohorts has already provided new important information concerning the natural history of b-cell autoimmunity and diabetes in early childhood. Based on our findings, we are proposing to follow these cohorts through the period of the highest risk of b-cell autoimmunity and to address the following goals: Specific Aims 1. To continue enrollment and follow-up, until the median age of 10yrs,of the existing cohorts of siblings/offspring and of newborns at high and moderate genetic risk detected through general population HLA-DR,DQ screening to: a) former define the incidence of b-cell autoimmunityby age,race/ethnicity, HLA-genotype, and family history of type 1 diabetes;b) formally evaluate candidate autoimmunity/diabetes risk factors available for all participants, e.g., early childhood diet, reported infections, and vaccination;and c) sustain this population laboratory for additional studies of type 1 diabetes and other autoimmune diseases. 2. To continue current and initiate new case-control studies, nested in the cohorts, of selected environmental and genetic risk factors for: a) b-cell autoimmunityand its persistence;and b) progression from b-cell autoimmunityto diabetes 3. Based on our findings, initiate intensive follow-up from birth of eighty highest risk children (HLA-DR3/4,DQB1 *0302 relatives) with monthly filter paper blood samples and weekly stool samples in addition to tri-monthly clinic visits. 4. To explore gene-environment interactions using combined approaches of case-control and case parent analyses. Our study is filling important gaps in the understanding of the events leading to type 1 diabetes in early childhood by providing the first unbiased population estimates of the incidence of b-cell autoimmunityand of the relative risks associated with candidate environmental and genetic factors.